Discrimination of wild and cultured european sea bass (Dicentrarchus labrax) using chemical and isotopic analyses

Recent legislation in the European Union (EC/2065/2001) requires that seafood must provide the consumer with information that describes geographical origin and production method. The present studies aimed to establish methods, based on chemical and stable isotopic analysis, that could reliably differentiate between wild and farmed European sea bass (Dicentrarchus labrax). The study measured fatty acid and isotopic compositions (delta13C and delta18O) of total flesh oil, delta15N of the glycerol/choline fraction, and compound-specific analysis of fatty acids (delta13C) by isotope ratio mass spectrometry. The sample set comprised 10 wild and 10 farmed sea bass from England (wild) and Scotland or Greece (farmed). Discrimination was achieved using fatty acid composition with 18:0, 18:2n-6, 20:4n-6, and 22:6n-3 providing the highest contributions for discrimination. Principal component analysis of the data set provided good discrimination between farmed and wild sea bass where factor 1 and factor 2 accounted for 60% of the variation in the data.     

Classification of gilthead sea bream (Sparus aurata) from 1H NMR lipid profiling combined with principal component and linear discriminant analysis

The combination of (1)H NMR fingerprinting of lipids from gilthead sea bream (Sparus aurata) with nonsupervised and supervised multivariate analysis was applied to differentiate wild and farmed fish and to classify farmed specimen according to their areas of production belonging to the Mediterranean basin. Principal component analysis (PCA) applied on processed (1)H NMR profiles made a clear distinction between wild and farmed samples. Linear discriminant analysis (LDA) allowed classification of samples according to the geographic origin, as well as for the wild and farmed status using both PCA scores and NMR data as variables. Variable selection for LDA was achieved with forward selection (stepwise) with a predefined 5% error level. The methods allowed the classification of 100% of the samples according to their wild and farmed status and 85-97% to geographic origin. Probabilistic neural network (PNN) analyses provided complementary means for the successful discrimination among classes investigated.     

Authentication of farmed and wild turbot (Psetta maxima) by fatty acid and isotopic analyses combined with chemometrics

Fatty acid composition and stable isotope ratios of carbon (delta(13)C) and nitrogen (delta(15)N) were determined in muscle tissue of turbot (Psetta maxima). The multivariate analysis of the data was performed to evaluate their utility in discriminating wild and farmed fish. Wild (n=30) and farmed (n=30) turbot of different geographical origins (Denmark, The Netherlands, and Spain) were sampled from March 2006 to February 2007. The application of linear discriminant analysis (LDA) and soft independent modeling of class analogy (SIMCA) to analytical data demonstrated the combination of fatty acids and isotopic measurements to be a promising method to discriminate between wild and farmed fish and between wild fish of different geographical origin. In particular, IRMS (Isotope Ratio Mass Spectrometry) alone did not permit us to separate completely farmed from wild samples, resulting in some overlaps between Danish wild and Spanish farmed turbot. On the other hand, fatty acids alone differentiated between farmed and wild samples by 18:2n-6 but were not able to distinguish between the two groups of wild turbot. When applying LDA isotope ratios, 18:2n-6, 18:3n-3, and 20:4n-6 fatty acids were decisive to distinguish farmed from wild turbot of different geographical origin, while delta(15)N, 18:2n-6, and 20:1n-11 were chosen to classify wild samples from different fishing zones. In both cases, 18:2n-6 and delta(15)N were determinant for classification purposes. We would like to emphasize that IRMS produces rapid results and could be the most promising technique to distinguish wild fish of different origin. Similarly, fatty acid composition could be used to easily distinguish farmed from wild samples.     

Classification of wines produced in specific regions by UV-visible spectroscopy combined with support vector machines

Discriminating wines according to their denomination of origin using cost-effective techniques is something that attracts the attention of different industrial sectors. In search of simplicity, direct UV-visible spectrophotometric techniques and different multivariate statistical techniques are used with admissible results to characterize wine produced in specific regions. However, most of the reported classification methods do not exploit all of the statistical relations in the investigated dataset and are inherently affected by the presence of outliers. The aim of this paper is to test novel classification methods such as support vector machines as a means of improving the classification rate when UV-visible spectrophotometric methods are used to discriminate wines. The advantages of such a discrimination tool are demonstrated when classification rates are compared for a large number of Spanish red and white wines and classification rates above 96% are achieved. The proposed methodology also enables the selection of the most relevant wavelengths for sample discrimination. The proposed methodology also enables the selection of the most relevant wavelengths for sample discrimination.     

Application of cryoprobe 1H nuclear magnetic resonance spectroscopy and multivariate analysis for the verification of corsican honey

Proton nuclear magnetic resonance spectroscopy ((1)H NMR) and multivariate analysis techniques have been used to classify honey into two groups by geographical origin. Honey from Corsica (Miel de Corse) was used as an example of a protected designation of origin product. Mathematical models were constructed to determine the feasibility of distinguishing between honey from Corsica and that from other geographical locations in Europe, using (1)H NMR spectroscopy. Honey from 10 different regions within five countries was analyzed. (1)H NMR spectra were used as input variables for projection to latent structures (PLS) followed by linear discriminant analysis (LDA) and genetic programming (GP). Models were generated using three methods, PLS-LDA, two-stage GP, and a combination of PLS and GP (PLS-GP). The PLS-GP model used variables selected by PLS for subsequent GP calculations. All models were generated using Venetian blind cross-validation. Overall classification rates for the discrimination of Corsican and non-Corsican honey of 75.8, 94.5, and 96.2% were determined using PLS-LDA, two-stage GP, and PLS-GP, respectively. The variables utilized by PLS-GP were related to their (1)H NMR chemical shifts, and this led to the identification of trigonelline in honey for the first time.     

Classification of wine and alcohol vinegar samples based on near-infrared spectroscopy. Feasibility study on the detection of adulterated vinegar samples

Near-infrared (NIR) spectroscopy was used to discriminate between wine vinegar (red or white) and alcohol vinegar. One orthogonal signal correction method (OSC) was applied on a set of 73 vinegar NIR spectra from both origins and artificial blends made in the laboratory in order to remove information unrelated to a specific chemical response (tartaric acid), which was selected due to its high discriminant ability to differentiate between wine vinegar and alcohol vinegar samples. These corrected NIR spectra, as well as raw NIR spectra and 14 physicochemical variables, were used to develop separate classification models using the potential functions method as a class-modeling technique. The aforementioned models were compared to evaluate the suitability of NIR spectroscopy as a rapid method for discriminating between vinegar origins. The transformation of vinegar NIR spectra by means of an orthogonal signal correction method prompted a notable improvement in the specificity of the constructed classification models. The classification model developed was then applied to artificial vinegar blends made in the laboratory to test its capacity to recognize adulterated vinegar samples.     

融合光谱形态特征的苹果霉心病检测方法

针对轻微霉心病和健康苹果光谱差异较小,致使基于可见/近红外特征光谱的检测方法对轻微霉心病检测准确率较低的问题。该研究将光谱形态特征与光谱特征融合的方法引入霉心病模型构建,建立了融合光谱形态特征的判别模型。以215个苹果可见/近红外光谱为样本,分析了不同预处理和特征提取组合对建模效果的影响,并完成了光谱特征的提取;分析健康果和霉心病苹果平均光谱的差异性,提取波峰、波谷等差异明显的光谱形态特征点,对比波段比、波段差和归一化强度差三类形态特征获取方法;最终建立光谱形态特征参数和光谱特征融合的苹果霉心病模型。试验结果表明,归一化预处理后提取的特征光谱和归一化强度差形态特征融合后模型判别准确率最高,在支持向量机模型中训练集、测试集判别准确率分别为98.6%和96.3%。特别是当发病程度小于10%时,该研究的判别模型准确率高于95%,表明通过融合光谱形态特征可以提升轻微病变霉心苹果的判别准确率。     

Assessment and monitoring of soil quality using near-infrared reflectance spectroscopy (NIRS)

Soil degradation processes have dramatically increased in their extent and intensity over the last decades. Progressively, actions have been taken in order to evaluate and reduce the major threats that have already wreaked havoc on soil conditions. Efficient and standardized monitoring of soil conditions is thus required but soil quality research is facing an important technological challenge because of the number of properties involved in soil quality. The objective of the present review is to examine critically the suitability of near-infrared reflectance spectroscopy (NIRS) as a tool for soil quality assessment. We first detail the soil quality-related parameters (chemical, physical and biological) that can be predicted with NIRS through laboratory measurements. The ability of imaging NIRS (airborne or satellite) for mapping a minimum data set of soil quality is also discussed. Then we review the most recent research using soil reflectance spectra as an integrated measure of soil quality, from global site classification to the prediction of specific soil quality indices. We conclude that imaging NIRS enables the direct mapping of some soil properties and soil threats, but that further developments to solve several technological limitations identified are needed before it can be used for soil quality assessment. The robustness of laboratory NIRS for soil quality assessment allows its implementation in soil monitoring networks. However, its routine use requires the development of international soil spectral libraries that should become a priority for soil quality research.     

基于近红外高光谱成像技术鉴别杂交稻品系

种子的筛选和鉴别是农业育种过程中的关键环节。该文基于近红外高光谱成像技术(874~1 734 nm)结合化学计量学方法以及图像处理技术实现杂交稻种的品系鉴别及可视化预测。采集了3类不同品系共2 700粒杂交水稻的高光谱图像,用SPXY算法,按照2∶1的比例划分建模集和预测集。基于水稻样本的光谱特征,采用主成分分析(PCA)方法初步探究3类样本的可分性。采用连续投影算法(SPA),提取出7个特征波长:985.08、1 106、1 203.55、1 399.04、1 463.19、1 601.81、1 645.82 nm。基于特征波长和全波段光谱,建立了偏最小二乘判别分析(PLS-DA)和支持向量机(SVM)模型。试验结果表明,所建模型判别效果较好,识别正确率均达到了90%以上,其中,SVM模型的判别效果优于PLS-DA模型,基于全谱的判别分析模型结果优于基于特征波长的判别模型。结合SPA-SVM校正模型和图像处理技术,生成样本预测伪彩图,可以直观的鉴别不同品系的水稻种子。结果表明,近红外高光谱成像技术可以实现杂交稻的品系识别及可视化预测,为农业育种过程中种子的快速筛选及鉴定提供了新思路。     

基于光谱和成像技术的作物养分生理信息快速检测研究进展

该文阐述了应用光谱和成像技术进行作物养分生理信息快速检测的主要研究进展和发展趋势。介绍了光谱和成像技术的基本原理、常用数据处理方法、建模方法和模型评价指标,重点总结了光谱和成像技术在5种常见农作物(水稻、小麦、油菜、玉米、大豆)的养分生理信息检测中的应用成果和研究进展(主要包括叶绿素类和氮素检测,病虫害、水分、杂草、重金属、农药胁迫诊断及产量预测等方面),分析了光谱和成像技术在作物生长信息检测的发展趋势。结果表明,光谱和成像技术能够快速无损获取作物养分生理信息,并能有效地对作物长势和逆境胁迫响应进行动态监测,对实现农业的精准化、数字化、信息化及智能化管理和作业具有重要意义。     

可见/近红外光谱分析秸秆-煤混燃物的秸秆含量

快速检测秸秆-煤混燃物对生物质混燃发电中补贴政策的制定具有重要意义。该研究采用可见/近红外光谱法定性判别秸秆、煤和秸秆-煤混燃物,定量分析秸秆-煤混燃物中秸秆含量。收集并制备秸秆样品80个（粒径小于80 mm）、煤样品9个（粒径小于10 mm）,制备秸秆质量分数为70%～99%的秸秆-煤混燃物样品120个（混燃物1）、秸秆分数含量为1%～30%的秸秆-煤混燃物样品120个（混燃物2）。使用FOSS NIRS DS 2500型光谱仪获取样品光谱。分别使用偏最小二乘判别法（PLS-DA）建立定性分析模型,使用改进的偏最小二乘法（MPLS）建立定量分析模型。结果显示,在秸秆和混燃物1之间进行判别,使用1100～2500 nm谱区,正确判别率为90.00%;在煤和混燃物2之间进行判别,使用400～2500 nm谱区,正确判别率为71.88%;定量分析混燃物1和混燃物2中秸秆含量,相对分析误差分别为2.32（400～2500 nm谱区）和1.48（400～1100 nm谱区）。研究结果表明,1100～2500 nm谱区较适合秸秆和混燃物1之间的判别,该谱区同样适合定量分析混燃物1中秸秆含量。400～1100 nm谱区较适合煤和混燃物2之间的判别,该谱区同样适合定量分析混燃物2中秸秆含量。可见/近红外光谱结合化学计量学是快速定性和定量分析大粒度秸秆-煤混燃物的可行方法。     

Chemometric discrimination among wild and cultured age-0 largemouth bass, black crappies, and white crappies based on fatty acid composition

The potential to distinguish juvenile wild from cultured fishes and to discriminate among juvenile fishes by species based on fatty acid composition was demonstrated. Statistical approaches to data evaluation included analysis of variance, correlation analysis, principal component analysis (PCA), and quadratic discriminant analysis (QDA). Differences were determined between wild and cultured fishes both within and between species and between hatcheries. Fatty acid compositions were compared among individual (not composited) specimens of wild and cultured, age-0, freshwater species: largemouth bass Micropterus salmoides, black crappies Pomoxis nigromaculatus, white crappies P. annularis, and black-nose crappies. Four fatty acids were investigated: linoleic acid (18:2n-6), linolenic acid (18:3n-3), arachidonic acid (20:4n-6), and docosahexaenoic acid (22:6n-3). Linoleic acid was the primary fatty acid used to differentiate juvenile wild from cultured fishes. Concentrations of linoleic acid were significantly different (p < 0.05) in cultured largemouth bass and black crappies from the wild counterparts. Linolenic acid concentrations were not significantly different (p < 0.05) between wild and cultured largemouth bass but were significantly different between wild and cultured black crappies. Wild largemouth bass contained higher concentrations of arachidonic acid than the cultured bass, and concentrations of docosahexaenoic acid were twice as high in wild black crappies than cultured black crappies. On the basis of four signature fatty acids, 90 of 91 juvenile fishes were correctly classified as wild or cultured; 32 of 37 wild juvenile fishes originating from the same reservoir were differentiated by species. Data from the training set were used to classify a test set of fishes as to species, source, or origin with 100% accuracy.     

融合面向对象与缨帽变换的湿地覆被类别遥感提取方法

为了有效提取湿地覆被类别遥感信息,该文基于国产环境星影像(HJ-CCD)和Landsat7遥感影像(ETM)提出了一种融合面向对象技术和缨帽变换的提取湿地覆被信息的方法,并对东洞庭湖区的湿地进行提取。遥感提取结果的总体精度90.02%,Kappa系数0.88,高于传统的分类方法分类的量化结果;获得的结果没有"椒盐现象"且比较紧致。试验结果表明融合面向对象和缨帽变换的方法能够有效的提取湿地覆被类别,精度高,效果好。研究结果为有效地利用遥感手段提取湿地覆被信息提供参考。     

田间作物杂草识别的最优遥感测量尺度

遥感分类识别精度受测量尺度的制约。为克服现有最优测量尺度选择方法存在的问题,该文提出一种基于光谱角匹配的最优测量尺度选择方法。该方法将每个像元的光谱看作其所属地物类别参考光谱叠加混合像元与光谱变异性的净效应的总和,计算不同空间分辨率下像元光谱与其所属地物类别参考光谱的光谱角,用以衡量混合像元与光谱变异性净效应的大小,当光谱角最小时说明混合像元与光谱变异性的净效应最小,此时的遥感测量尺度即为最优尺度,并在1幅实例数据中实现了该方法,利用基于光谱角匹配的尺度选择方法得到了最优遥感测量尺度,通过试验证明在该尺度下进行分类识别时精度优于比其更大或更小的尺度,验证了本研究提出的最优空间分辨率选择方法的可靠性。将该实例数据中的目标地理实体对象化,从理论上分析了目标对象的面积和形状指数与最优遥感测量尺度之间的关系。该研究为田间作物杂草遥感识别提供了一种有效的最优测量尺度选择方法,可为当前变量作业中田间数据获取工作提供参考,对于推动遥感测量尺度选择研究也具有积极意义。     

光源光斑直径对苹果霉心病近红外检测的影响

为实现中早期霉心病苹果的有效剔除以提高苹果的整体品质，该研究利用近红外光谱技术对苹果霉心病进行快速无损检测，从光谱和分类模型两方面探究光源光斑直径对苹果霉心病检测的影响。在30、50 及70 mm光源光斑直径条件下采集了苹果样本的透射光谱，分析不同光源光斑直径下健康苹果和霉心病苹果的光谱差异，然后应用支持向量机（support vector machines，SVM）和粒子群算法优化-最小二乘支持向量机（particle swarm optimization-least squares support vector machine，PSO-LSSVM）方法建立苹果霉心病的分类模型，并对不同光源光斑直径下的分类模型性能进行对比。在此基础上，采用竞争自适应重加权采样（competitive adaptive reweighted sampling, CARS）方法筛选特征波长变量并建立分类模型。研究结果表明，30 mm光源光斑直径对苹果霉心病的检测效果最好，建立的SVM和PSO-LSSVM分类模型性能均最优。30 mm光源光斑直径下，最优PSO-LSSVM模型的预测集的灵敏度、特异度和正确率分别为89.5%、95.5%和92.7%。CARS-PSO-LSSVM分类模型性能比全波段的分类模型性能略有下降，预测集的灵敏度、特异度和正确率分别为89.5%、90.9%和90.2%，但建模变量数仅占原波长变量数的4.2%，有效地简化了分类模型。该研究为苹果霉心病的快速无损高精度检测提供技术支撑。     

Estimating hydrogen cyanide in forage sorghum ( Sorghum bicolor ) by near-infrared spectroscopy

Hydrogen cyanide (HCN) is a toxic chemical that can potentially cause mild to severe reactions in animals when grazing forage sorghum. Developing technologies to monitor the level of HCN in the growing crop would benefit graziers, so that they can move cattle into paddocks with acceptable levels of HCN. In this study, we developed near-infrared spectroscopy (NIRS) calibrations to estimate HCN in forage sorghum and hay. The full spectral NIRS range (400-2498 nm) was used as well as specific spectral ranges within the full spectral range, i.e., visible (400-750 nm), shortwave (800-1100 nm) and near-infrared (NIR) (1100-2498 nm). Using the full spectrum approach and partial least-squares (PLS), the calibration produced a coefficient of determination (R(2)) = 0.838 and standard error of cross-validation (SECV) = 0.040%, while the validation set had a R(2) = 0.824 with a low standard error of prediction (SEP = 0.047%). When using a multiple linear regression (MLR) approach, the best model (NIR spectra) produced a R(2) = 0.847 and standard error of calibration (SEC) = 0.050% and a R(2) = 0.829 and SEP = 0.057% for the validation set. The MLR models built from these spectral regions all used nine wavelengths. Two specific wavelengths 2034 and 2458 nm were of interest, with the former associated with C═O carbonyl stretch and the latter associated with C-N-C stretching. The most accurate PLS and MLR models produced a ratio of standard error of prediction to standard deviation of 3.4 and 3.0, respectively, suggesting that the calibrations could be used for screening breeding material. The results indicated that it should be feasible to develop calibrations using PLS or MLR models for a number of users, including breeding programs to screen for genotypes with low HCN, as well as graziers to monitor crop status to help with grazing efficiency.     

Mass spectrometry based sensor strategies for the authentication of oysters according to geographical origin

This study was undertaken to investigate the relevance of using the pyrolysis-MS (Py-MS) technique to discriminate the production area of oysters harvested over two years and to assess from the data of the second year of harvest the potential of an alternative MS-based technique, the solid phase microextraction-MS (SPME-MS), to perform this discrimination. Oysters were harvested in various areas of France, and models of discrimination according to harvest season were built from Py-MS fingerprints and from virtual SPME-MS fingerprints obtained by summing the mass spectra generated by the SPME-GC-MS system. The treatment of the Py-MS data by a 21-12-3 artificial neural networks led to a correct classification of only 89.2% of the oyster samples according to shoreline. The misclassifications thus did not allow use of the Py-MS technique as a relevant tool for authentication of oyster origin. The assessment of the potential of the virtual SPME-MS fingerprints to discriminate the production area of oysters was undertaken on a part of the sample set. The virtual SPME-MS data were pretreated according to two methods, filtering of raw data (FRD) and comprehensive combinatory standard correction (CCSC), a recently developed chemometric method used for the correction of instrumental signal drifts in MS systems. The results obtained with the virtual SPME-MS fingerprints are promising because this technique, when the data were pretreated by the CCSC method, led to a successful discrimination of the oyster samples not only according to shoreline but also according to production region. This study confirms that an efficient correction method (CCSC) of instrumental drifts can considerably increase the discriminative information contained in the volatile fraction of food products.     

基于关联规则面向对象的海岸带海水养殖模式遥感识别

针对目前海水养殖模式遥感识别中的效率低,"同物异谱"、"异物同谱"和"椒盐"噪声等问题,该文研究了关联规则分类和面向对象相结合的养殖模式遥感识别方法,通过不同养殖模式的对象分割和关联规则的自动和智能获取,来构建海水养殖模式分类器。以高分一号PMS1卫星影像为数据源,把不同养殖模式对象的光谱、空间形态和纹理特征及其关联关系作为事务数据,使用Apriori算法挖掘类别作为后件的强规则,对粤东柘林湾养殖核心区内4种海水养殖模式(池塘养殖、网箱养殖、滩涂插养、浮筏吊养)水面信息进行提取。结果表明:基于关联规则面向对象的海水养殖模式分类精度能达到88.65%,比K-近邻法面向对象法精度提高了14.38个百分点,比关联规则挖掘分类法精度提高了12.16个百分点。关联规则分类和面向对象结合方法拓宽了传统逻辑推理分类方法中获取信息的途径,使分类更加自动化和智能化,且分类精度得到显著提高,可以成为海岸带海水养殖复杂模式识别的有效支持手段。     

利用近红外高光谱图像技术快速鉴别西瓜种子品种

为了研究采用近红外高光谱图像技术对西瓜种子品种快速无损鉴别的可行性,该文采用近红外高光谱图像技术,通过提取西瓜种子的光谱反射率,结合Savitzky-Golay (SG)平滑算法,经验模态分解算法（empirical mode decomposition,EMD）和小波分析（wavelet transform,WT）对提取出的光谱数据进行去除噪声处理,采用连续投影算法（successive projections algorithm,SPA）和遗传-偏最小二乘法（genetic algorithm-partial least squares, GA-PLS）进行特征波长选择。基于全波段光谱建立了偏最小二乘判别分析（partial least squares-discriminant analysis,PLS-DA）,基于特征波长建立了反向传播神经网络（back-propagation neural network,BP NN)判别模型和极限学习机（extreme learning machine,ELM）判别模型。试验结果表明,基于特征波长的BPNN模型和ELM模型的结果优于基于全部波长的PLS-DA模型,基于SG预处理光谱提取的特征波长建立的ELM模型取得最优的判别效果,建模集和预测集的判别正确率均为100%。结果表明应用近红外高光谱成像技术对西瓜种子品种鉴别是可行的,为西瓜种子的品种快速鉴别提供了一种新方法。     

柑桔黄龙病近红外光谱无损检测

为探讨快速无损检测柑桔黄龙病的可行性,应用近红外光谱技术结合机器学习方法进行研究。在4000~9000cm-1光谱范围内,采集黄龙病、缺素和健康3类叶片样本的近红外光谱。采用一阶导数、平滑和多元散色校正组合的光谱预处理方法,消除光谱的基线漂移和散射效应。分别对偏最小二乘判别模型(PLS-DA)的主成分因子数和最小二乘支持向量机(LS-SVM)的输入变量数量、核函数类型及其参数进行了优化,建立了PLS-DA和LS-SVM模型。采用预测集样本,评价模型的预测能力,经比较,采用11个主成分得分向量为输入、线性核函数和惩罚因子为2.25的LS-SVM模型预测效果最佳,模型误判率为0。结果表明采用近红外光谱技术结合最小二乘支持向量机进行柑桔黄龙病无损检测是可行的。